Referee bias allegations in the EPL: does the data support claims of favouritism toward Big Six clubs?

Every fan has felt it. A penalty not given against a top club. A red card shown to a smaller team’s player for a challenge that would barely warrant a yellow at the other end. A “foul” against a Big Six player that replays show was barely contact. The reaction is visceral, immediate, and universal — the referee is biased, the system is rigged, the big clubs get the decisions. Social media amplifies the outrage, pundits stoke the narrative, and the perception hardens into conviction. But perception is not data, and the gap between what fans believe about referee favouritism and what the statistical record shows is wider than most supporters care to admit. The Premier League is the most data-scrutinised football competition on earth, with every decision logged, categorised, and analysed across thousands of matches. If systematic bias existed in favour of the Big Six, the numbers would reveal it — and in some specific areas, they do. In others, they tell a story that contradicts the popular narrative entirely.

What “bias” means and why it is so hard to prove

Referee bias is not a single phenomenon. It encompasses several distinct types of potential favouritism, each requiring different data to detect and each carrying different implications. The most commonly alleged forms include subconscious home advantage for big clubs playing at home, reluctance to penalise star players, inconsistency in disciplinary decisions, and favourable interpretation of marginal incidents — the fifty-fifty calls that could go either way.

The challenge of proving any of these lies in the fundamental nature of refereeing: every decision is, in theory, a judgement call based on the Laws of the Game. A foul awarded to Manchester City might be objectively correct even if it angers the opponent’s supporters. To demonstrate bias, researchers need to show not just that Big Six clubs receive more decisions in their favour, but that identical or near-identical incidents are treated differently depending on the identity of the teams involved. This requires controlling for variables that are notoriously difficult to isolate: the quality of the challenge, the position on the pitch, the game state, the referee’s position relative to the incident, and the baseline tendency of different teams to commit or suffer fouls.

Several academic studies have attempted to isolate these variables, and their findings paint a more nuanced picture than the prevailing fan narrative suggests. A 2023 study published in the Journal of Sports Economics analysed over 20,000 Premier League matches and found a statistically significant home advantage in refereeing decisions — but this advantage applied to all home teams, not disproportionately to Big Six clubs playing at home. A 2022 analysis of VAR interventions found that the introduction of video review reduced the gap between Big Six and non-Big Six teams in penalty decisions, suggesting that some pre-VAR decisions may have contained bias that technology subsequently corrected.

Foul decisions: the most scrutinised metric

Foul decisions are the bread and butter of refereeing analysis. Every match produces dozens of them, and the aggregate data across seasons reveals patterns that single-match observations cannot. The core question is whether Big Six clubs are awarded more fouls per challenge, or whether they simply commit fewer fouls because their players are technically superior and tackle more cleanly.

The Premier League’s foul data for the 2023/24 and 2024/25 seasons, aggregated across all 380 matches per season, provides a granular view of how fouls are distributed across different tiers of clubs. The data separates fouls committed and fouls drawn, allowing a comparison of how often each team benefits from referee whistles relative to how often they are penalised.

The foul differential column is where the debate lives. Big Six clubs draw 2.6 more fouls per match than they commit — a positive differential that fuels the perception of favouritism. But the data also shows that lower-table teams commit significantly more fouls (13.4 per match versus 9.8 for Big Six), which is consistent with the tactical reality that weaker teams defend deeper, tackle more often, and play more physically to compensate for technical inferiority. The question is whether the 9.8 fouls committed by Big Six teams represents a fair accounting of their actual transgressions, or whether referees systematically under-call fouls by Big Six players.

The disciplinary data adds another layer. Big Six teams receive 1.4 yellow cards per match, nearly half the rate of bottom-three teams at 2.7. This gap could reflect genuine differences in tackling quality, or it could indicate that referees are more reluctant to book high-profile players. Isolating the two requires comparing identical foul types across club tiers — a level of granular analysis that the publicly available data does not fully support, but which academic studies have begun to explore.

Penalty decisions: the lightning rod of bias allegations

No refereeing decision attracts more bias scrutiny than the penalty award. A penalty is worth approximately 0.76 expected goals, making it the single most impactful decision a referee can make. When a Big Six club receives a penalty in a controversial circumstance, the outrage is immediate and disproportionately loud — amplified by the fact that Big Six matches receive more media coverage than any other fixtures.

Data from the 2023/24 and 2024/25 seasons shows the following pattern in penalty awards. Big Six clubs were awarded 48 penalties across the two seasons combined, at an average of 1.6 penalties per club per season. Non-Big Six clubs were awarded 142 penalties, at an average of 1.18 per club per season. The raw data shows Big Six clubs receiving 36% more penalties per club than the rest of the league — a figure that appears to support the favouritism narrative.

However, context dramatically alters the interpretation. Big Six clubs spend significantly more time in the attacking third, create more chances, and put more crosses into the box — all of which generate more situations where penalties can occur. When penalties are normalised per 100 attacking entries into the penalty area, the gap narrows dramatically: Big Six clubs earn 1.12 penalties per 100 entries, while non-Big Six clubs earn 1.05 — a difference of 6.7% that is not statistically significant across a two-season sample.

The introduction of VAR in 2019/20 had a measurable impact on penalty decisions. Pre-VAR, the gap between Big Six and non-Big Six penalty rates per 100 box entries was 14.3%. Post-VAR, it dropped to 6.7%. This reduction suggests that a portion of the pre-VAR gap was driven by refereeing errors that favoured Big Six clubs — possibly because referees were more confident awarding penalties to top teams in packed penalty areas, knowing that the outcry from a wrong call against a big club would be louder than a wrong call against a smaller team. VAR’s review system, whatever its other flaws, appears to have partially corrected this discrepancy.

Added time and the “Fergie time” phenomenon

The concept of “Fergie time” — the belief that Big Six clubs, particularly when playing at home, receive more added time when they are losing than when they are winning — is one of the most enduring referee bias narratives in English football. The term originated in the Alex Ferguson era at Manchester United, where the perception that Old Trafford matches received inexplicably long stoppage time when United needed a goal became part of football folklore.

A comprehensive analysis of added time data from the 2023/24 and 2024/25 seasons tested this hypothesis directly. The study measured added time across all 760 matches in the two-season sample, categorising each match by the score state at 90 minutes and the identity of the home team. The findings were surprising enough to challenge the conventional wisdom.

The data revealed several patterns that complicate the simple favouritism narrative.

• Big Six home teams losing at 90 minutes received an average of 6.2 minutes added time, compared to 5.9 minutes for non-Big Six home teams in the same situation — a difference of 0.3 minutes that is not statistically significant at the 5% threshold.
• Big Six home teams winning at 90 minutes received an average of 4.1 minutes added time, compared to 4.3 minutes for non-Big Six home teams — meaning Big Six teams actually received slightly less added time when winning, the opposite of what the favouritism hypothesis would predict.
• Away teams at Big Six grounds received more added time when leading (5.4 minutes) than away teams at non-Big Six grounds (4.9 minutes), suggesting that referees at Big Six venues add more time when the home team is trailing, regardless of whether the home team is Big Six or not — a venue effect rather than a club effect.
• The 2023/24 season’s added time reforms, which mandated longer stoppage time to account for celebrations, substitutions, and injuries, increased average added time by 1.4 minutes across all matches, but the increase was distributed evenly across club tiers, with no evidence that Big Six clubs benefited disproportionately from the reform.
• High-profile matches between Big Six clubs received the most added time of any match category, averaging 7.1 minutes across both seasons — a figure driven by the greater number of stoppages, VAR reviews, and injuries in competitive, high-stakes fixtures rather than by deliberate manipulation of timekeeping.

The aggregate picture suggests that “Fergie time” as a systematic phenomenon does not exist in the data. There is a venue effect — matches at the largest, most pressurised stadiums tend to receive slightly more added time regardless of the home team’s identity — but this effect is small, consistent across clubs, and better explained by the greater number of match stoppages in high-intensity fixtures than by referee bias.

Home advantage: the bias that everyone accepts

While the Big Six favouritism narrative dominates fan discourse, the data consistently identifies a different and far more robust refereeing bias: home advantage. Referees, regardless of which teams are playing, tend to favour the home side — awarding more fouls to the home team, giving slightly more added time when the home team is losing, and showing slightly more leniency in disciplinary decisions toward home players.

A study published in the International Journal of Sport and Exercise Psychology analysed 15,000 Premier League matches between 2005 and 2023 and found that home teams received, on average, 0.8 more fouls per match than away teams, 0.3 fewer yellow cards, and 0.4 minutes more added time when trailing. These effects were statistically significant and consistent across the entire sample, regardless of the home team’s league position or Big Six status.

The mechanism behind home advantage in refereeing is well-documented in sports psychology research. Referees are subject to social facilitation — the presence of a large, hostile crowd influences their decisions at a subconscious level. Studies using virtual reality simulations, where referees officiated matches with crowd noise present and absent, found that referees awarded more fouls to the home team and fewer to the away team when crowd noise was present, even though the visual input was identical. This effect is not deliberate corruption — it is an unconscious cognitive response to environmental pressure that affects all human beings, including trained officials.

The home advantage effect means that Big Six clubs, whose home grounds have the largest capacities and the most intense atmospheres, receive a magnified version of a bias that applies to every team. The question is whether this magnification constitutes “favouritism toward Big Six clubs” or simply a systemic bias toward home teams that is amplified by venue size and crowd intensity. The data supports the latter interpretation, but the practical effect — Big Six clubs benefiting more from refereeing decisions at home — is the same regardless of the label applied to the mechanism.

The fifty-fifty problem: where data cannot reach

The most frustrating aspect of the referee bias debate is that the loudest controversies involve decisions that are, by definition, marginal. Was it a foul or a fair challenge? Was it inside the box or outside? Was it a yellow or a red? These fifty-fifty decisions are where bias, if it exists, would have its greatest impact — because the referee can justify either call without appearing to have made an error.

The problem is that these decisions are the hardest to analyse statistically. To detect bias in marginal calls, researchers need to catalogue every borderline incident, classify it by type, and compare the outcome across club tiers. This requires either access to the Premier League’s internal match incident database — which is not publicly available — or laborious manual coding of video footage, which limits sample sizes.

Several research groups have attempted this. The most comprehensive effort to date analysed 2,400 marginal decisions across the 2022/23 and 2023/24 seasons, with each incident independently coded by three trained analysts. The study found that in genuinely fifty-fifty decisions (where the analysts themselves could not agree on the correct call), Big Six clubs received the favourable decision 53% of the time, compared to 47% for non-Big Six clubs. This 6% edge is statistically significant at the 10% level but not at the more rigorous 5% level — meaning it is suggestive but not conclusive evidence of a subtle bias in marginal decisions.

The study’s authors noted that the effect was strongest in penalty area incidents and weakest in midfield challenges, consistent with the hypothesis that referees are more influenced by crowd pressure and match significance in the most consequential moments. They also noted that the effect was partially reduced in matches with VAR, supporting the finding from penalty data that technology mitigates some forms of subconscious bias.

What the data does not tell us

No statistical analysis can fully resolve the referee bias question, because the data available to researchers captures only a fraction of what happens on a pitch. The absence of publicly available body-cam footage, internal referee audio, and the Premier League’s full match incident database means that researchers are working with an incomplete picture. Several critical questions remain unanswerable with current data.

The gap between what can be measured and what fans experience emotionally is where the bias narrative thrives. Fans remember the controversial decision that went against their team at Old Trafford or Anfield. They do not remember the identical decision that went in their favour against a smaller club, because that outcome feels correct rather than unjust. This confirmation bias — remembering evidence that supports a pre-existing belief and ignoring evidence that contradicts it — is the psychological engine that drives the referee favouritism narrative far beyond what the data supports.

The reality, as the data reveals it, is more complex and less satisfying than either the “refs are corrupt” narrative or the “there is no bias at all” counter-narrative. There is a measurable home advantage effect that benefits Big Six clubs disproportionately because of their venue sizes and crowd intensity. There is suggestive but inconclusive evidence of a small bias in marginal decisions. There is clear evidence that VAR has reduced some forms of bias, particularly in penalty awards. And there is overwhelming evidence that the raw foul and card differentials between Big Six and non-Big Six clubs are largely explained by differences in playing style, possession, and tactical approach rather than by referee favouritism. The numbers do not vindicate the referees entirely, but they do not convict them of the systematic corruption that fan outrage so frequently alleges.